The purpose of this edited volume is to provide a comprehensive overview on the fundamentals of deep learning, introduce the widely-used learning architectures and algorithms, present its latest theoretical progress, discuss the most popular deep learning platforms and data sets, and describe how many deep learning methodologies have brought great breakthroughs in various applications of text, image, video, speech and audio processing.

Deep learning (DL) has been widely considered as the next generation of machine learning methodology. DL attracts much attention and also achieves great success in pattern recognition, computer vision, data mining, and knowledge discovery due to its great capability in learning high-level abstract features from vast amount of data. This new book will not only attempt to provide a general roadmap or guidance to the current deep learning methodologies, but also present the challenges and envision new perspectives which may lead to further breakthroughs in this field.

This book will serve as a useful reference for senior (undergraduate or graduate) students in computer science, statistics, electrical engineering, as well as others interested in studying or exploring the potential of exploiting deep learning algorithms. It will also be of special interest to researchers in the area of AI, pattern recognition, machine learning and related areas, alongside engineers interested in applying deep learning models in existing or new practical applications.

本书的目的是全面介绍深度学习的基本原理，介绍广泛使用的学习架构和算法，介绍其最新的理论进展，讨论最流行的深度学习平台和数据集，并描述许多深度学习方法如何在文本、图像、视频、语音和音频处理的各种应用中带来巨大突破。

深度学习（DL）已被广泛认为是下一代的机器学习方法。由于深度学习在从大量数据中学习高层次抽象特征方面的巨大能力，它吸引了很多人的注意，也在模式识别、计算机视觉、数据挖掘和知识发现方面取得了巨大的成功。这本新书不仅试图为当前的深度学习方法提供一个总体路线图或指导，而且还提出了挑战并设想了新的视角，这可能会导致这一领域的进一步突破。

本书将作为计算机科学、统计学、电气工程专业的高年级（本科或研究生）学生，以及其他对研究或探索利用深度学习算法潜力感兴趣的人的有用参考。人工智能、模式识别、机器学习和相关领域的研究人员，以及对在现有或新的实际应用中应用深度学习模型感兴趣的工程师，也会对这本书特别感兴趣。